0.3 release?

[Roland McGrath]

Marcus asked me about the subject of a hurd-0.3 release.
I guess random people ask about this fairly often.

I don't really know quite what it means or how it would really matter.
Directly getting the development code from CVS is the only thing that
really serves anyone wanting to seriously hack the Hurd.  Debian/Hurd is
the only thing really serving someone who wants to use the Hurd, and as far
as I'm concerned Marcus is chief arbiter of what's ready for prime time.

I don't see what making a GNU hurd-0.3 release would accomplish, other than
to have an announcement on info-gnu and maybe draw some attention.  I also
don't see any harm done by making a release, nor any reason to fritter over
what goes into it or wait for any particular hurdle before making one.
It's not like there is a stable hurd-0.2 release we are superceding.

Perhaps we should make a hurd-0.3 release very soon, without waiting for
anything but immediately pending bug fixes.  It has sure been a long time
since 0.2, and there's a certain logic in making a "stable" release
containing the innumerable random fixes and enhancements done in that time,
before embarking on substantial new features, protocol changes,
implementation restructuring, etc.

I don't really have any preference one way or the other.  Nowadays, we can
always just refer to a particular date's cvs state, or particular debian
packge versions, so people don't need releases just to have access to the
code like they did back in the second millennium.